Thread
Commits
-
doc: Add missing index entries and fix title formatting in pg_buffercache docs.
- a8acfb133cf9 18.0 landed
- 086b9a33aafe 19 (unreleased) landed
-
doc: pg_buffercache documentation wordsmithing
- 8eede2c7200f 19 (unreleased) cited
-
Minor Improvements to pg_buffercache documentation
Fujii Masao <masao.fujii@oss.nttdata.com> — 2025-07-02T16:30:13Z
Hi, While reading commit 8eede2c7200 and the pg_buffercache docs, I noticed a few minor issues: - There are missing index entries for pg_buffercache_numa() and pg_buffercache_usage_counts(), even though other functions have them. - The section titles for pg_buffercache_evict_relation and pg_buffercache_evict_all are missing parentheses, unlike the others. The attached patch addresses these points by: - Adding index entries for pg_buffercache_numa() and pg_buffercache_usage_counts(). - Making the function section titles consistent by including parentheses where they were missing. Also, a quick note on pg_buffercache_numa_pages(): although there's an index entry for pg_buffercache_pages(), I didn't add one for pg_buffercache_numa_pages() because both functions are typically accessed via views, not directly. Regarding those functions, it seems enough to have index entries for the views. So, it might make more sense to remove the existing index entry for pg_buffercache_pages() for consistency. BTW, all current index entries point to the top of the pg_buffercache docs. Would it be better if each entry pointed directly to its relevant section using a zone attribute like this? - <indexterm> + <indexterm zone="pgbuffercache-summary"> <primary>pg_buffercache_summary</primary> </indexterm> Thought? Regards, -- Fujii Masao NTT DATA Japan Corporation -
Re: Minor Improvements to pg_buffercache documentation
Michael Paquier <michael@paquier.xyz> — 2025-07-22T05:10:32Z
On Thu, Jul 03, 2025 at 01:30:13AM +0900, Fujii Masao wrote: > BTW, all current index entries point to the top of the pg_buffercache > docs. Would it be better if each entry pointed directly to its relevant > section using a zone attribute like this? > > - <indexterm> > + <indexterm zone="pgbuffercache-summary"> > <primary>pg_buffercache_summary</primary> > </indexterm> That would be smarter than the current statu-quo of the page, especially if this page grows more in size in the future. The other inconsistencies are good catches, by the way! -- Michael
-
Re: Minor Improvements to pg_buffercache documentation
Fujii Masao <masao.fujii@gmail.com> — 2025-07-24T02:56:36Z
On Tue, Jul 22, 2025 at 2:10 PM Michael Paquier <michael@paquier.xyz> wrote: > > On Thu, Jul 03, 2025 at 01:30:13AM +0900, Fujii Masao wrote: > > BTW, all current index entries point to the top of the pg_buffercache > > docs. Would it be better if each entry pointed directly to its relevant > > section using a zone attribute like this? > > > > - <indexterm> > > + <indexterm zone="pgbuffercache-summary"> > > <primary>pg_buffercache_summary</primary> > > </indexterm> > > That would be smarter than the current statu-quo of the page, > especially if this page grows more in size in the future. The summary descriptions for each function and view are currently at the top of the docs. If we update the index entries to point directly to the relevant sections instead of the top, it might also make sense to move those summaries into their respective sections. I’ll consider this later. > The other inconsistencies are good catches, by the way! Thanks for the review! I've pushed the patch. Regards, -- Fujii Masao